Remy Reference/Models
Models
349 models · 14 types · refreshed daily
The managed model catalog behind the Agent SDK: every language, image, video, audio, embedding, and reranking model you can reach with a model override, each with its live price, limits, and capabilities. It is refreshed daily from the platform, so the catalog tracks what the SDK can actually call today. You do not install this. You point the SDK's generation actions at it.
Language
Text-in, text-out models for chat, reasoning, tool use, and coding, selected by model override.
130 models
Image generation
Models that generate and edit images from a prompt or a source image.
71 models
Video generation
Models that generate video from text, a still image, or a source clip.
49 models
Embeddings
Models that turn text into vectors for semantic search and retrieval.
20 models
Text to speech
Models that turn text into spoken audio, with a choice of named voices.
12 models
Vision
Models that read images and answer questions about them in text.
67 models
3D generation
Models that generate 3D assets from text or an image.
11 models
Transcription
Models that turn spoken audio into text.
10 models
Reranking
Models that reorder search results by relevance to a query.
9 models
Realtime
Speech-to-speech models for live, interruptible voice conversations.
8 models
Lipsync
Models that sync a face in a video to a track of speech.
8 models
Music generation
Models that compose music from a text description.
5 models
Document extraction
Models that pull structured text out of documents and images.
3 models
Video analysis
Models that watch a video and answer questions about it in text.
2 models